Diane asked in the "Comments" session what my thoughts were so far on 2x2s. Also, there has been some discussion on various LiveJournal agility blogs about the pros and cons of the method. I thought I would address them here.
Overall, I am EXTREMELY happy with the 2x2 method. I cannot imagine using any other method to teach a dog -- any dog -- weaves in the future. It is easy. It is equipment-minimal. The beginning stages are small-space friendly. The whole method is small-yard friendly. There is nothing to fade. The dog learns that weaving is this super awesome game that s/he plays with momma and "oh boy, I can't wait to go weave!!!"
I fully intend to retrain Tessie's weave poles using this method, starting this week. (We have a trial this weekend, and I didn't want to confuse her or let her think 2 poles was acceptable for the time being.) She is 10 years old, been doing agility since she was 4, and I hate retraining anything on her because time may be short and I don't like causing confusion. HOWEVER, there's nothing confusing about this method!
Now I'd like to address some of the concerns I've read about...
"It's not non-Border-Collie friendly!"
I strongly disagree. Here's a Kelpie learning the method, an English Cocker Spaniel trained with the method, a sporting-type dog (pointer of some sort?) as a demo dog in the beginning stages, a Coton de Tulear in her fourth session, and a Portugese Water Dog workin' on them, too. Of course, Strata is a Sheltie, so he counts, because Shelties are definitely NOT mini BCs. I'd also like to point out the YouTube videos that Susan Garrett put up herself, such as this one from her Contacts & Weaves Camp and this one that was a promo for the DVD including a couple of Labs, a Giant Schnauzer, and an unknown mini dog (Pap?).
"I don't like my dog failing that much."
If your dog is exceeding the 80% fail rate you DO need to make it easier. Note "easier" does not mean changing positions. "Easier" means that you go back to the 2-pole stage and work a little more there. I did not need to do this with Strata, but I am sure I may need to use it in the future on another dog. Alternatively, you can see in some videos that Strata gets "overexcited" at running in circles and I take him by the collar and lead him to *the spot where I was standing*, wait for him to focus on the poles, and release him. Sometimes it is not a matter of understanding, just a matter of getting super hyped-up by all the running and tugging.
People often cite the "3 strikes, you're out" rule, which is fine for trained behaviors and handling manuevers (I use this all the time). For example, if you are free shaping a dog to back up and he decides to start offering nose touches, do you reward that? No. You wait for the dog to offer something resembling backing up, even if it's just a paw movement or weight shift, and reward that and then move from there. You might decrease your criteria a smidge for a few repetitions, then proceed to where you were before. But I don't care if my dog offers 500 nose touches -- we're working on backing up!
If you are worried about a dog shutting down due to "failure", you need to free shape more behaviors. A couple of blog-followers have asked me about this and I suggest you have free-shaped (NOT lured, NOT captured) a minimum of 5 behaviors before attempting the 2x2 training. Some easy examples to get you started: limping, the dog wipes its face with its paw, backing up, lowering its head ("are you sad?"), lifting a rear leg, a 2o/2o behavior on stairs (even if it's not your desired contact behavior).
"You have to use a NRM."
I haven't watched the DVD, so I cannot say for sure that she does not REQUIRE you to use a NRM, but I don't think that's the case. I do not use a NRM with my dogs any longer (it wreaked havoc on poor Tessie) and I can tell you that you do not have to use one to succeed. I know that in the ebook, it is suggested that you might want to use one, but it is not forced upon you. Also, I believe the NRM is for the end stages of the behavior, i.e. the dog is weaving a 6 pole set but starts trying to enter with the first pole on the right, for example. I would never use a NRM in the shaping stage.
"It only works for toy-motivated dogs."
The ebook talks about when/where to use cookies. If you really have a dog with no toy drive you could use a Manners Minder, throw extremely large/visible cookies (freeze dried chicken hunks work well on most footing), or try a bait bag or food tube instead. Tessie is completely NOT toy motivated at all, so when I retrain her I will be able to speak on this topic with a tiny bit more authority.
If you have any more questions or would like me to elaborate, please do not hesitate to ask! I enjoy reading the responses from all of you, even if it's just "Keep up the good work!" or "Your dog is sooo cute!" Positive reinforcement works for bloggers, too. ;)
Subscribe to:
Post Comments (Atom)
4 comments:
Great post. Thanks so much. Diana
I completely agree. NRM isn't necessary, and reserved for end-stage behavior. And you can use the Manners Minder---i've been using mine. Great post!
Re positive reinforcement for bloggers: I think your sessions look good! One thing you might want to watch is that Strata's head is turning back to you sometimes for the reward. I have the same problem with a thrown reward with my dog too. I think it's something everyone battles with thrown rewards.
You may also wish to try to skim the reward closer to the ground, maybe a little more like bowling, so the dog doesn't learn to weave and then jump up. Of course that may not happen, but I'm always surprised by the number of dogs I see that weave with a very high head position, because they've been lured and rewarded from above. So you may with to change your throw just a little bit.
When I originally trained 2x2s, I did not understand how to work all the positions on the arc, esp. the more difficult ones. Looks like you're doing just fine with that part. Good job; I enjoyed watching.
I know my throws are terrible! :( I have broken things inside from misplaced "click-and-throws". I've practiced, but the unpredictability of his toy doesn't really help. I also think part of it is that he looks back when he hears the click (and now with the loss of audio in my videos, you, the viewer, can't tell that). He knows the reward is on its way.
The good thing is that the "look back" doesn't interfere with his weaving. If he was interrupting his behavior chain to ask "am I right? how about now? is this it? oh my goodness, what about this?" and constantly turning his head, I'd be more concerned. But he only does it when he's heard the click or is damn well sure that he's right. Good observation though!
Post a Comment